The spatial organization of innovation strategies in the Eindhoven region : A study on the socio-spatial implications of innovative networks in the high-tech sector of the Eindhoven region

1

The spatial organization of innovation

strategies in the Eindhoven region

A study on the socio-spatial implications of innovative networks in the high-tech

sector of the Eindhoven region

Stefan Behlen

May 2014

2

The spatial organization of innovation

strategies in the Eindhoven region

A study on the socio-spatial implications of innovative networks in the high-tech

sector of the Eindhoven region

Master thesis Economic Geography

in combination with a graduate internship at the SRE

Radboud University Nijmegen

Faculty of Management Sciences

Master Economic Geography

Author:

Stefan Behlen

0823937

Date:

May 2014

Courtesy of:

Supervising lecturer:

prof. dr. Frans Boekema

Internship supervisor:

drs. Thomas Voncken

3

Preface

With completion of the research project presented in this master thesis the master program "Economic Geography" at the Radboud University is coming to an end. Meanwhile, it is some time since I started writing this thesis. During the writing of this thesis, I have followed an internship at the SRE (Samenwerkingsverband Regio Eindhoven) for eight months. The role of public entities on cluster development had caught my interest, so I was very pleased that I was received by the SRE with open arms, where I was immersed among several tasks of the Economic Affairs department. This

experience helped me to develop this interest into a subject for the thesis. Partly due to the freedom and in addition the guidance I was given at the SRE, I could focus on doing my research. Because of this internship, I was able to work with a bulk of information on the spatial organization of innovation networks in the region Eindhoven.

Concerning this thesis I owe a lot to my supervisor Thomas Voncken. He helped me with the

collection of data, and directed me by focusing on the study. The many conversations inspired me to keep faith in myself. He advised me to pick up experiences as much as possible during my internship that are important for me, so I could develop my professionalism and personality.

I would also like to thank prof. dr. Frans Boekema, my thesis supervisor at the Radboud University. He appeared to be an inspiring mentor to me. Again and again, he made sure that I could bring this thesis a step further towards a final product. His comments and constructive attitude towards the subject in combination with emphasizing on my own responsibilities as a "researcher", I experienced as very pleasant. Because of the long duration of the research and writing of this thesis it was sometimes difficult to stay motivated. Thanks to his sharp and constructive feedback he made sure that I could resume this study with vim and vigour every time.

Finally, I would like to thank the other colleagues of the SRE for the pleasant cooperation and the respondents for their contribution to the research. Without them, this outcome would not have been possible.

Stefan Behlen May 2014

4

Summary

According to many recent conceptual developments in the innovation and economic geography literature, networks can be seen as qualitative relationships which are reflected in

inter-organizational knowledge relationships. It has been widely argued that these personal networks across organizations are critical for innovation processes, in particular in high-technology sectors where the external sourcing of knowledge through networks has been identified as an important mechanism. These insights suggest that learning within a cluster is a process of social interaction among individuals rather than a rationally organized process within and between firms and that, consequently, learning should be understood from the social context in which it takes place.

However, the role of institutions on local scale to innovation and learning remain underexplored. So if the public sector like the SRE wants to stimulate innovation and cluster development through incentive funds and by operating as an accelerator and connector in the collaboration with

stakeholders, they must be aware of the current and future spatial strategies of key actors within a cluster as well as the way these strategies are constituted.Therefore they need to be provided with information about the role of space, as well as the role of knowledge (including their spatial

implications) that constitute the structure of innovation strategies of key actors involved in innovative activities in the high-tech sector.

In the literature, there is a variety of conceptions to be found on spatial theories on innovation. Among  others,  ideas  of  the  theoretical  perspectives  ‘regional  innovation  systems’  and  ‘learning   regions’  have  gained  increasing  influence  in  regional  studies  and  policies  during  the  last  one  and a half decade. There are currently two trends of development of the original theories to be discovered. One trend is the diversification of the notion of proximity made by Boschma (2005). He shows that not only the spatial concentration, but also other social variables play a role in reinforcing a cluster. There can be four other forms of proximity be distinguished besides spatial proximity: cognitive proximity, social proximity, organizational proximity and institutional proximity. Another trend is the assumption made by Asheim et al. (2005a), whom state that policy implications can be derived from spatial innovation strategies, stemming from socio-spatial implications of knowledge. He discusses how different industries rely on different knowledge bases (analytical, synthetic and symbolic) for activities that are most central to their competitiveness. These two approaches form the theoretical basis of the spatial organization of innovation strategies, which can be a: stand alone, local buzz or global pipelines strategy.

Therefore, the novel contribution of this study is to take a perspective on the spatial organization of innovation strategies of key actors within the HTSM sector, in order to provide the public sector (SRE)

5 policy recommendations concerning enhancement of innovation and cluster development in the region. This is done on the basis of a qualitative case study, where the data on the main

characteristics of the spatial context in which knowledge transfers and networks in the BRE arise will be obtained on the basis of documents and other studies and the data on the spatial organization of innovation strategies of key actors within the HTSM sector will be obtained by depth-interviews, resulting in an improved insight that lead to policy recommendations concerning enhancement of innovation and cluster development in the region.

The analysis shows that behind the economic success of the Brainport region lies a centuries -old history of entrepreneurship, a decades-long tradition of regional cooperation and ten years of organizational experience in collective strengthening of open innovation processes. Particularly the economic structure of OEMs and suppliers, many spin-offs and knowledge institutions are considered by the respondents as most important location advantages for innovative technological firms.

Concerning the five forms of proximity, I can conclude that some factors are more critical than others in the networks. The nature of the type of knowledge that is particularly developed in this region - in this case it concerns high-tech knowledge which means production is characterized by low volume and high precision – is concerned with a process of prototyping and testing which involves much uncertainty. It seems that this process particularly benefits from a high degree of physical and social proximity, and to a lesser extent cognitive, organizational and institutional proximity. As for

innovation strategies many firms use the knowledge they received from both the external resources (local and global) in their innovation process. Next to the contacts within the cluster, non regional relations seem to be very important to create new knowledge. The combination of contacts on both scales can strengthen the competitive position of the region. In that sense it is more accurate to speak of a place-based development in (international) economic networks than  of  a  ‘learning  region’.   This is in line with the philosophy behind Smart Specialization, where the focus is placed on the strategic potential of a regional area to stimulate collaboration - between regions, sectors and levels in order to increase efficiency in research and innovation.

Conventional wisdom is that R&D has been globalized, so innovation strategies are increasingly focused on global pipelines. However, this study challenges such understanding and provides a more nuanced conclusion. Multinationals, and other institutions deriving on analytical based knowledge make use of global pipelines but have their R&D centers in the BRE due to path-dependency of the region. They play a peripheral role due to the advantages of local buzz. The implications from these findings are substantial. While global pipeline strategies seems rampant, you cannot outsource or geographically separate certain processes of creating innovations. Global pipeline strategies may take place, but only if a component or functionality is specifically defined and separated. The

6 innovation process, including the engineers and managers who create innovations, have to be in proximity, and the role of local buzz has not diminished.

The history of the Eindhoven region shows that targeted policy pays off. A concrete proposal is accelerating processes in which political decision of the SRE is involved, because their role is to monitor the economic interest of the region and this process can be slowed by the individual

municipalities in the region, whom mainly guard their own interest. In addition, the SRE should focus on the economic activities within the core of the region, without blocking economic activities that are related to other geographical areas to prevent a lock-in effect. It is important for the SRE that they provide particular support to SMEs in the form of carrying risks in innovation processes as they have fewer resources. However, the SRE should not interfere too much with the course of economic processes itself as the role of a governmental institution is primarily preconditional. Meanwhile they should particularly deliver financial stimulation, motivation and directing a more flexible labor market. This can be achieved on the basis of innovation instruments. In addition, they can play a crucial role in creating alliances and other connections by focusing on open innovation.

7

Table of Contents

Chapter 1: Introduction

9

1.1 Dynamics in the region

9

1.2 Research problem

10

1.3 Research objective

12

1.4 Relevance

13

1.5 Research design

14

Chapter 2: Theory

16

2.1 A development in theories on the relation between space and learning

16

2.2 Spatial concentration

18

2.3 The diversification of the notion of proximity

20

2.3.1 Geographical proximity

21

2.3.2 Cognitive proximity

21

2.3.3 Social proximity

22

2.3.4 Organizational proximity

23

2.3.5 Institutional proximity

24

2.4 The spatial implications of knowledge

25

2.4.1 Industries drawing on different knowledge bases

25

2.4.2 Analytical knowledge base

26

2.4.3 Synthetic knowledge base

26

2.4.4 Symbolic knowledge base

27

2.5 Conceptual model

27

2.6 Innovation strategies

28

2.7 Reflection of the notion of proximity on innovation strategies

29

2.8 Reflection of the implications of knowledge on innovation strategies

31

2.9 Implications for policy

32

Chapter 3: Methodology

35

3.1 Research strategy

35

3.2 Research material

36

3.2.1 Interviews

36

3.2.2 Data-analysis

39

Chapter 4: Spatial concentration in the Brainport region

40

4.1 Introduction

40

4.2 Path-dependency

40

4.3 Geographical factors

42

8

4.5 Specialized labor

46

4.6 Specialized suppliers

48

4.7 Knowledge spillovers

49

4.8 Conclusion

50

Chapter 5: Notion of proximity and the implications of knowledge transfers

52

5.1 Introduction

52

5.2

Geographical proximity

52

5.3 Cognitive proximity

54

5.4 Social proximity

56

5.5 Organizational proximity

57

5.6 Institutional proximity

60

5.7 Analytical & synthetic knowledge base

61

5.8 Conclusion

64

Chapter 6: Global pipelines or local buzz?

65

6.1 Introduction

65

6.2 Innovation projects

65

6.3 Global pipelines

66

6.4 Local buzz

70

6.5 Conclusion

73

Chapter 7: Conclusion

74

7.1 Operational structure of networks

74

7.2 Innovation projects

75

7.3 In short

77

Chapter 8: Recommendations

78

References

81

Appendices

85

9

1. Introduction

1.1 Dynamics in the region

The high-tech region Eindhoven is undergoing a metamorphosis. On the foundations of the manufacturing industry and knowledge economy is rapidly developing a successful knowledge industry. Firms, knowledge institutions and government in Southeast Brabant are working together to further develop Brainport Region Eindhoven (BRE), already one of the strongest economic regions of the Netherlands (Brainport Development, 2013). In order to retain that position, the region has to anticipate on future developments. The regional corporation consisting of 21 municipalities – known as Samenwerkingsverband Regio Eindhoven (SRE) - realizes for quite some time now that a dynamic on several domains is taking place which have a major impact on the development of the high-tech region. From an economic point of view there is a growing realization that the region needs smart, sustainable and inclusive growth (European Commission, 2011). This goes along with some societal challenges that play an increasing role in the coming years such as a shortage of raw materials, aging, climate change and a growing demand for food and energy (Brainport Commission, 2010). Brainport 2020 - a strategy program that has to constitute the region Southeast Netherlands into a leading position in the international knowledge economy - is a response to these developments by focusing on innovation in the top sectors: High Tech Systems & Materials, Chemicals, including Lifetech, Food and the Creative Industries (Brainport Commission, 2010). This strategy program can be considered as the guideline for the regional agenda 2011-2014 that is established by SRE.

The sector that stands out and has a big influence on the economic developments in the BRE is High Tech Systems & Materials (HTMS). This is one of those sectors of which a large part is concentrated in and around Eindhoven. This sector is seen as a key area of BRE (Brainport Development, 2013). This is due to the combination of business activity and knowledge in that region, which in a sustainable growth market can cope with the global competition. Of course, innovation is of decisive importance. It has commonly been assumed that innovation is the main driver for economic growth. Innovation strengthens the competitiveness of countries as well as sectors and individual companies (Porter, 1985). For the scope of this study, I will focus on innovation processes within the HTSM sector because; the region has a prominent position in this sector; innovation in this sector can cope with future sustainable developments; and can push the region into the desired leading position. The HTSM sector includes a number of closely interlinked manufacturing industries: machinery and systems industries, automotive, aerospace and materials including steel. According to High-Tech Systems Platform (n.d.), this sector generated in 2008 a turnover of 74 billion Euros, with an added value of 20 billion Euros. This industry is capital intensive, and jointly invest more than 2.3 billion

10 Euros a year in its own research and development. Some companies export more than 90% of their production, other invest up to 20% of their turnover on R&D.

Although there is a good basis for innovation in the HTSM sector, it is in the interest of the

competitive position of the region that it focuses more on advanced technological solutions to social issues and sustainable developments because it creates a long-term selling market. The public sector is able to make a greater contribution to the innovative power of the region by focusing on the opportunities for new economic drivers. For this purpose, the SRE is authorized to deploy the Regional Incentive Fund (of 5 million a year) to promote innovation and cluster development (SRE, 2011). The SRE has notably a supporting role in achieving this goal. This is evident from the following tasks to realize the goals in the technology domain:

- Strengthen research and knowledge in the region.

- Supporting international knowledge and knowledge networks.

- Extension of the financing of knowledge by the (national) government.

Such a transition on innovative development requires a common vision and agenda for all

stakeholders within a cluster. The CPB (2010) concludes that clusters, along with trained and highly qualified people, are the driving forces of the future economic strength. In these nodes, knowledge, capital, talent, infrastructure and residential and working environment are present and allows companies to compete globally. In particular, the presence of public R&D infrastructure is crucial for the appeal to businesses and people. Investing in top sectors like HTSM, will therefore precipitate in spatial clusters, where they render the most.

1.2 Research problem

As mentioned, BRE wants to strengthen the region to retain their top position. This has lead to a Regional Agenda that has the explicit ambition to strengthen the ecosystem of the core area of BRE. Part of that program is that the public sector (such as SRE) has to stimulate innovation and cluster development through incentive funds and by operating as an accelerator and connector in the collaboration with stakeholders. The role of the public sector in driving innovation is currently too weak. This is, inter alia, reflected in the current low public spending on R&D infrastructure, which makes the region vulnerable. When companies close down or remove their R&D departments, gaps will fall in the system which on their turn shall damage the top clusters. A strong and accessible public knowledge infrastructure has the ability to bind, to encourage and attract private R&D. This is vital for product development, export and for the establishment and business conditions (Brainport Commission, 2010). A strong innovation infrastructure funded by the public sector can indeed

11 generate new markets and opportunities for regional business. This can be done by reinforcing existing top clusters, connecting them, have them working together with other regional networks and by looking for new sustainable markets. SRE expects to achieve this by involving the following actors: firms, learning institutions and government. They need to cooperate in order to reach advanced innovative processes.

As slightly shines through here, innovation in the region has a strong link with cooperation and thus networks. According to many recent conceptual developments in the innovation and economic geography literature, networks can be seen as qualitative relationships which are reflected in inter-organizational knowledge relationships (Malmberg & Maskell, 2002). It has been widely argued that these personal networks across organizations are critical for innovation processes, in particular in high-technology sectors where the external sourcing of knowledge through networks has been identified as an important mechanism (Powell et al. 1996). These insights suggest that learning within a cluster is a process of social interaction among individuals rather than a rationally organized

process within and between firms and that, consequently, learning should be understood from the social context in which it takes place (Rutten & Boekema, 2012). However, the role of institutions on local scale to innovation and learning remain underexplored (Lorentzen, 2005). In order for SRE to be able to serve as a supporting party in the development of the regional knowledge infrastructure, they must be aware of the current and future spatial strategies of key actors within a cluster as well as the way these strategies are constituted.Therefore they need to be provided with information about the role of space, as well as the role of knowledge (including their spatial implications) that constitute the structure of innovation strategies of key actors involved in innovative activities in the high-tech sector.

Within this thesis, I want to draw the attention on the high tech innovation cluster and networks in the BRE. The influence of institutions of different scales on the interactions and their spatial implications has not yet been studied intensively. Much is still unclear on the conditions for private companies to share knowledge with other companies, and on the type of knowledge they are willing to share, even within a high-tech region where most empirical work is focused on (Lorentzen, 2005). This may partly be due to the fact that even within in the sector of HTSM there are signs of much diversity: A common tripartite typology, also used in the context of large companies, is that of: 1. frontrunners, 2. followers and 3. laggards. Each has its own characteristics and susceptibility for stimulation by external stakeholders. However, this distinction is not very specific. Companies can take on different positions and strategies on different issues at different moments in time. Innovative behavior is dynamic by nature and influenced by a multitude of factors inside and outside the region. Therefore, a view on various influences to the spatial structure of networks, their local implications,

12 and consequently strategies, needs to be conceptualized. This dissertation aims to contribute to this insight on HTSM specific factors surrounding advanced innovation processes.

1.3 Research objective

The novel contribution of this study is to take a perspective on the spatial organization of innovation strategies of key actors within the HTSM sector, in order to provide the public sector (SRE) policy recommendations concerning enhancement of innovation and cluster development in the region. In the literature, there is a variety of conceptions to be found on spatial theories on innovation. Among others,  ideas  of  the  theoretical  perspectives  ‘regional  innovation  systems’  and  ‘learning  regions’  have   gained increasing influence in regional studies and policies during the last one and a half decade (Lorentzen, 2005). These perspective are based on a series of assumptions related to the scale and dynamic of innovation.These include place and region (Paasi, 2002), embeddedness (Granovetter, 1985), and innovation as a system in space and scale (Bathelt, 2003). According to Lorentzen (2005), there are currently two trends of development of the original theories to be discovered. One trend is the diversification of the notion of proximity (Boschma, 2005). The other is the inclusion of the global or extralocal linkages in the models (Bathelt et al. 2004). However for the conceptualization of this study I will partly adopt the assumption made by Asheim et al. (2005a), whom state that policy implications can be derived from spatial innovation strategies, stemming from socio-spatial

implications of knowledge. Bathelt & Glückler (2003) express this as the type of knowledge which is needed in the innovation process, as well as the degree to which this knowledge is localized (or sticky). Along with the meaning and role of space to innovative activities, these approaches form the theoretical basis of the spatial organization of innovation strategies. The choice for these two approaches is based on the fact that the role of space and the role of knowledge seem to be interconnected and highly relevant for the high tech sector, as well that it contains both qualitative as quantitative characteristics of networks.

This study will take the geographic area (BRE) as starting point for the emergence, structuring and evolution of innovative activities. This does not mean that a cluster is conceived as a geographical restricted  agglomeration  and  that  their  activities  don’t  transcend  these boundaries. Instead, for some innovation processes or projects it is the quality of the technological expertise of the partners that matters, rather than a mere geographical closeness (Preissl & Solimene, 2003). Further this study intends to contribute to the understanding of the main characteristics of knowledge in the high tech sector in the BRE. By studying these social innovation networks it becomes clear why precisely this region is so strong in the HTSM sector and how the public sector such as SRE can strengthen the

13 innovative region that it retains the position of a top region. The main research question of this study is formulated as follow:

In what way can SRE support innovation and cluster development in the Eindhoven region, looking at the way the spatial organization of innovation strategies of key actors within the HTSM sector is constituted?

Sub research questions involve the following:

1. Do innovative high-tech firms spatially concentrate in Brainport Eindhoven? 2. What is the role of space and knowledge on innovation networks?

3. How are the notion of proximity, and the implications of knowledge reflected in innovation strategies of key actors in Brainport Eindhoven?

1.4 Relevance

Scientific relevance

The objective of this paper is to take a perspective on the spatial organization of innovation strategies of key actors within the region. The particular spatial organization depends on both qualitative and quantitative characteristics of the knowledge networks which are needed for innovative processes. According to Sugden et al. (2005) a concern for such qualitative relationships and their evolution among the actors of clusters does not seem to have received very much attention. They state that this topic has not been pursued much further in recent literature, and such issues are clearly not the focus of existing methodologies. Sugden et al. (2005) and other critics, state that the majority of existing methodologies tend to centre on relatively superficial features. This study tends to focus on qualitative affecting factors of networks within an cluster. Especially the emphasis on the social aspects concerning personal relationships in innovative networks is a significant contribution to the scientific theory. This is argued by Doloreux & Parto (2005), whom state that that the interactions between the actors in regional innovation systems have been insufficiently explored while the institutional context of these interactions has been by and large overlooked. As a result, the validity of the recommendations for innovation policy making, based on the current analyses of regional innovation systems is somewhat questionable. This model will be build on existing

theoretical perspectives concerning the role of space and knowledge. According to Asheim et al. (2005a), different scholars have studied these concepts in various settings, resulting in the awareness that all kind of factors affect spatial organization of knowledge networks. But often these potentially relevant factors have not been included in the analysis and has not been systematically studied. In addition there is little empirical evidence on factors influencing the adjustment between government

14 and business concerning innovation policy and strategy within a regional context. This study

attempts to fill that lack of empirical evidence by doing field research on the spatial organization of innovation strategies of key actors within a HTSM context so that policy can be adjusted accordingly.

Societal relevance

The greatest societal benefits gained by the insights from this study on the understanding of the effect of knowledge networks on innovation strategies are achieved from the societal need to embed advanced innovation processes adequately. The High Tech top sector makes an important

contribution to advanced innovation processes and products. So when the development of

innovation in this sector can be supported it will help to improve the local living standards. This is of particular importance for practitioners from governments, trade associations and other intermediary organizations who have a stake in achieving the strategy program for the BRE, which includes

responding to the societal challenges that play an increasing role in the coming years. Commissioned by the SRE this study will in particular lead to recommendations on the field of advisory and

supportive functions, as they struggle with questions how to stimulate innovation and cluster development. This struggle is among others characterized by a lack of insight in the way the spatial organization of innovation strategies is constituted.

1.5 Research design

In this paragraph, the research model will be drafted. This shows how the study will be

operationalized and how the central concepts are linked together. As in Part A of the model below is shown schematically, the theoretical framework will form the research perspective to approach the empirical reality. Primary sources from the literature are formed by two key theoretical perspectives, known  as:  ‘regional innovation  systems’  and  ‘smart  specialization’. In Part B it is clear that the spatial organization of innovation strategies of key actors in the HTSM sector is analyzed. In Part C, the step from analysis to testing is denoted. The analysis will be reviewed on a basis of an assessment, divided into two dimensions. This is done by means of a case study that has to deliver the desired data and knowledge. Part D shows how the theory testing study will then lead to the desired outcome, namely, improved theoretical insight on innovation strategies, resulting in policy recommendations. With qualitative research, based on interviews, the new insights will be obtained. In the final phase of the study the main research question is answered and recommendations are made.

15

16

2. Theory

Innovation is closely associated with processes of knowledge-creation, the development of new technologies, and the effects of technological change, especially in a spatial perspective. According to Bathelt & Glückler (2003), many traditional concepts in economics and geography fail to properly understand the process of generating new products and processes and introducing innovations to established markets. Technological change is either viewed as a given, being external to the models used, or portrayed as a predictable outcome of a linear research process which follows a controlled sequence of research and development stages. More recent evolutionary interpretations, such as those of Storper (1997), have realized that the process of generating new technologies has to be conceptualized with care. The creation of new technologies is viewed as an interactive social process, taking place in a particular spatial dimension. One could therefore say that knowledge transfers are characterized by its spatial implications.

2.1 A development in theories on the relation between space and learning

The interest in local factors behind development, innovation and competitiveness have its roots in earlier developments. One could say it started with the early industrial districts theory. This theory points at the economic benefits of co-location (personal contacts and information flows) of firms (Marshall,  1890).  Many  years  later,  Porter  (1990)  developed  the  concept  ‘cluster’. A  ‘cluster’  can  be   defined as a geographically proximate group of interconnected companies, service providers and associated institutions in a particular field, linked by externalities that connect the constituent industries, such as common technologies, skills, knowledge, and purchased inputs. The conception of clusters was first conceived at the national level, later on the relevant geographic unit covered regions.  Another  development  is  the  approach  of  ‘new  industrial  spaces’  (Scott  &  Storper,  1992),   which focuses on a district with a collection of small firms that have the same specialization and draw in common external resources. Together the firms may achieve economies of scale as well as

economies of scope. According to Lorentzen (2005), new production technology has enabled small firms to be extremely flexible internally as well as in relation to each other, meaning that the district as a whole is innovative, highly responsive to market changes, and therefore competitive on the global market. A more recently developed approach - that focuses directly on innovation - is the ‘French  innovative  milieu’  in  which  the  milieu  is  the  entrepreneur and innovator (Ache, 2004). This approach argues that co-location, personal relations and networks are important because they determine the local capacity to respond to competitive pressures. An innovative milieu is a coherent system of firms and organization which through communication and interaction create and share a common culture and approach to problems and situations (Lorentzen, 2005). The milieu is able to

17 search for new knowledge outside the milieu and to translate it into knowledge understandable to the milieu (Ache, 2004). The local environment is more important than the individual firm.

Insights  from  the  above  mentioned  theories  are  integrated  in  the  ‘regional  innovation  system’   approach. The basis of these theories is pointed on the interaction of companies, knowledge institutions and government, in relation to the production, creation and distribution of knowledge (Lorentzen, 2005). Together these organizations constitute an infrastructure (a system of innovation) on which the performance of an economy depends on. A basic assumption is that spatial proximity between economic actors is advantageous. It facilitates the exchange of information considering innovation is an interactive process among economic actors. With that assumption in mind, the ‘region’  is  seen  as  an  efficient  level  for  communication  and  knowledge  sharing.  Communication  and   knowledge sharing is mostly facilitated by participation in local networks and associations.

The existence of localized learning networks explain the emergence of new and competitive regions. The engagement of firms, institutions and public sector in processes of self-organized interactive learning is determined by the social capital of such networks (Lorentzen, 2005). This continues process can be seen as  the  starting  point  of  the  theories  on  the  ‘learning  region’.  The  innovation   system and regional learning approaches - which are based on a series of assumptions related to the scale and dynamic of innovation - have been the starting point of many studies on spatial dimensions of innovation.

More recently there are noises of conceptualizations of the relation between space and learning that goes beyond the learning region. For example, Rutten & Boekema (2013) argue that an individual perspective, where spatial range and spatial embeddedness of innovation networks are

conceptualized as networks variables, may be an alternative to the perspective of the learning region. They state that learning is a process of interaction between individuals whom are affected by their embeddedness in organizations and society. Such social variables affect learning, and

subsequently innovation becomes a network rather than a regional characteristic.

Smart Specialization

A concrete example of such conceptualization that goes beyond the  ‘learning  region’  which  currently is much discussed is: Smart Specialization. This concept has been incorporated by the European Commission in their innovation policy design, in order to promote the efficient and effective use of public investment in research (European Commission). Their goal is to boost regional innovation in

18 order to achieve economic growth and prosperity, by enabling regions to focus on their strengths (i.e. Regional competitiveness). According to Midtkandal & Sörvik (2012), Smart Specialization or RIS3 (Research and Innovation strategies for Smart Specialization) is a strategic approach to economic development through targeted support for research and innovation. It involves a process of developing a vision, identifying the place-based areas of greatest strategic potential, developing multi-stakeholder governance mechanisms, setting strategic priorities and using smart policies to maximize the knowledge-based development potential of a region, regardless of whether it is strong or weak, high-tech or low-tech. Over the past few years, the concept of Smart Specialization has been diffused at surprisingly rapid pace among European regions. The latest economic crises, in combination with demographic challenges, climate changes and increased global competition, have increased European attention to research, innovation and entrepreneurship. To implement the strategy  ‘Europe  2020’  for  smart,  sustainable  and  inclusive  growth,  Smart  Specialization  strategies   were introduced as a way to increase efficiency in research and innovation investments by

integrating policy areas, applying a broad definition of innovations and stimulating collaboration - between regions, sectors and levels (Lindqvist, 2012).

In these smart specialization policies, regional competitiveness is introduced as an essential element. Regions are to be considered as a place-based development in a networked competitive setting and not a purely local-assets setting. Place-based smart specialization strategies integrate regional profiling with the implementation of a regional development smart specialization strategy where policy is targeted at the key priorities and strengths on which policy actions are able to build, as well as the weaknesses, bottlenecks or missing links which need to be rectified, in a network of economic relations (Thissen et al. 2013).

2.2 Spatial concentration

In both, regional economic and economic-geographical literature it is often assumed that firms benefit from the existence of near-by, and thus physical proximity of firms and other organizations like universities. Since the late 80s, the importance of innovation and knowledge for the

competitiveness and development of the region is strongly emphasized. Spatial proximity between organizations would facilitate the exchange of knowledge (Boschma, 2005). A frequently cited explanation is that the spatial proximity between organizations in concentration regions leads to many formal and informal contacts, which increases the likelihood of knowledge transfers. Ever since the publication of the work of Marshall (1890) it is recognized that firms located in a region where other firms and related organizations in the same sector are concentrated can benefit from spatial economies of scale. Marshall identifies three advantages: specialized labor, specialized suppliers and

19 knowledge spillovers. The regions where sectors concentrate and where firms benefit from these advantages are referred to as clusters. For decades, it is emphasized that firms in such clusters benefit from lower transaction costs. As a result of technological developments, the transport costs declined sharply in recent decades and the communication possibilities greatly increased. As a result, the cost of contracting and maintaining contacts strongly declined in time and money. It was

expected that this spatial concentration of industries would decrease. But the continued - and even emerging - regional concentrations of industries around the world prove otherwise (Weterings & Ponds, 2007). Many empirical studies on this phenomenon show that there is a positive correlation between regional concentration of industries and the number of innovations. In the nineties, the focus on the explanation of the advantages of spatial concentration shifts from lower transaction costs to benefits of knowledge spillovers (Glaeser, 1998). Knowledge spillovers concern the learning processes that occur due to intentional or unintentional utilization of knowledge by other

organizations than where it is originally produced. This can occur in various ways, for example via interactions between organizations, labor mobility or observing behavior (Vicente & Suire, 2007).

Agglomeration advantages

In general, the benefits of physical concentration of organizations can be distinguished into two types of agglomeration advantages: localization and urbanization advantages (Malmberg & Maskell, 2002). In the case of localization advantages, firms benefit from an establishment in a region where firms, operating in the same sector are concentrated. Examples include the creation of a specialized labor market or suppliers. Urbanization advantages are the benefits that a region offers, where many firms are concentrated. In other words the larger cities. In more densely populated regions you will often find a better developed infrastructure or concentration of entities such universities and research laboratories. Since the 90s, also static and dynamic agglomeration advantages are distinguished. Static agglomeration advantages are the cost benefits which can be provided by an establishment nearby other organizations. Dynamic agglomeration advantages are caused by the accumulation of knowledge in areas where firms are concentrated. Over time, relationships arise between firms which makes it easier for them to exchange knowledge. This would stimulate learning processes in those regions. The knowledge spillovers can both occur in regions where firms, specialized in the same sector are concentrated and in larger cities. According to Malmberg & Maskell (2002), knowledge from one sector often can be successfully applied in other sectors. Therefore, the most knowledge spillovers occur in regions where different types of firms are located.

20

2.3 The diversification of the notion of proximity

The model designed by Bathelt, Malmberg & Maskell (figure 4) shows that not only the spatial concentration, but also other social variables play a role in reinforcing a cluster. According to

Boschma (2005) there can be four other forms of proximity be distinguished besides spatial proximity: cognitive proximity, social proximity, organizational proximity and institutional proximity.

The diversification of the notion of proximity seems paradoxical. According to Boschma & Frenken (2009) this can be argued as follows. A high degree of proximity has a positive impact on the creation of a possible collaboration between actors. In other words, actors tend to look for partnerships over a small distance. At the same time it seems that a high degree of proximity does not automatically lead to more innovative outcomes stemming from these partnerships. Even more, Boschma & Frenken (2009) argue that an excess of proximity between actors can have a negative effect on these outcomes.The opposite is also true. A low degree of proximity may impede a possible cooperation between actors, while this is not necessarily detrimental to the innovative outcomes of a

partnership. So the proximity paradox deals with the fact that proximity influences the choice of partners positively, while it may be detrimental for the outcomes of entered relationships when it comes to interactive learning, knowledge sharing and development, and ultimately the creation of innovations. This means that there may be an optimal proximity that increases the probability of co-operation between actors and simultaneously stimulates the innovative outcomes. In the following sections, an overview will be given of the assumed effect of the degree of proximity to the creation of partnerships between actors with a view to interactive learning and knowledge.

Figure 2: Local buzz and global Pipelines Source: Bathelt & Malmberg & Maskell (2004)

21 2.3.1 Geographical proximity

Geographical or spatial proximity is probably the most well known and studied form of proximity. Geographical proximity refers to the spatial or physical distance between two economic actors, both in absolute and relative terms. Small physical distance literally brings people together, which

stimulates information- and tacit1 knowledge transfers. The more distance between the actors, the more difficult it is to transfer tacit knowledge. Even the transfer of codified2 knowledge can be complicated by greater physical distance because it is assumed that the interpretation and

assimilation of this kind of knowledge requires a certain degree of tacit knowledge and thus requires spatial proximity.

Maskell (2001) concluded that the co-location of similar activities makes a positive contribution to the dissemination of knowledge and innovations. The theories of "industrialized clusters and agglomerations" are largely based on the positive externalities of spatial proximity. Although

geographical proximity encourages interaction and cooperation it is not a prerequisite for interactive learning and innovation (Malecki & Oinas, 1999). Despite geographical proximity can make a positive contribution to interactive learning and innovation by facilitating the creation of relationships, an excess of geographical proximity can be obstructive. There is a risk of spatial lock-in, whereby actors are too much inward focused and become too dependent, making it hard to pick up and value external opportunities.

2.3.2 Cognitive proximity

Cognitive proximity is the extent to which acting persons or organizations can identify, assess and exploit  each  other’s  individual  knowledge  framework or knowledge base. The knowledge framework and knowledge base varies between different organizations and because knowledge development and learning to a large extent depends on the combination of diverse, complementary skills and knowledge of different people within and between organizations, it is important to bring these together to stimulate exchange (Nooteboom, 2000). With cognitive proximity is thus meant, the extent to which two individuals or organizations share the same knowledge base. Because of the idiosyncratic and tacit feature of much knowledge it is not enough to only have access to new

knowledge. It is necessary that the absorptive capacity enables the acting persons or organizations to identify, assess and exploit new knowledge (Boschma, 2005). In other words, it requires a certain degree of cognitive proximity between actors in order to absorb new knowledge, which means that

1 _{Tacit knowledge is often gained from experiences en requires personal interaction.} 2

22 their own cognitive basis, or knowledge framework, is close enough to the new knowledge in order to communicate, understand and process it successfully.

On the other hand, too much cognitive proximity may cause a prejudice on learning and innovation. First, because knowledge development is dependent on bringing together different, complementary and not exactly the same kinds of knowledge. Second is that too much cognitive proximity may cause a lock-in. A third reason why too much cognitive proximity may be unfavorable lies in the risk of involuntary and uncontrollable spillover effects. When individuals or organizations function within the same technological discipline or within the same operational area, the possibilities to exchange complementary skills and knowledge are small, while there is a risk that unintended spillovers are created regarding the current knowledge base. Nooteboom (2000) summarizes the importance of cognitive distance and proximity as follows: "A tradeoff needs to be made between cognitive distance, for the sake of novelty, and cognitive proximity, for the sake of efficient absorption. Information is useless if it is not new, it is useful bit useless if it is so new that it cannot be understood."

2.3.3 Social proximity

Economic activities do not stand on their own, but take place in a broader context and should also be considered such. To a large extent economic activities are embedded in a social context, social relationships influence economic processes. In contrast to the neo-classical economic literature, the ‘embeddedness’ literature suggest that "the more socially embedded are the relationships of a firm, the more interactive learning and the better its (innovative) performance" (Boschma, 2005, p66). Social proximity is therefore defined as socially embedded relationships between individuals at the micro level, where the anchor is based on trust on the basis of friendship, kinship and experience. Indeed, within such relationships is also information available about possible potential partners, which increases the chance of entering into new innovative partnerships or networks (Boschma & Frenken, 2010).

The idea that social proximity encourages interactive learning and innovation is due to the fact that social relationships, based on trust, facilitates the transfer of tacit knowledge (Malmberg & Maskell 2002). Boschma & Frenken (2010) explain that an important aspect of social proximity that

particularly emerges in entering new network relations is called "closure". This refers to the emergence of a new relationship in which two actors are introduced by a third, meaning that both actors already have a relationship and that they therefore to some extent can trust each other.

23 Too much or too little social proximity may also undermine the creation or establishment of

(innovative) relationships between individuals or organizations. On the one hand, too much social proximity may undermine the innovative capacities of individual or organizational relationships by an overload of confidence. First, because there may be negative consequences to social interaction in "a world with calculating actors, in markets where technologies and policies continually change in conditions of uncertainty, and where opportunism is a common attitude" (Boschma, 2005, p.66). In addition, when social relations are too much anchored, or if there is too much involvement,

members of a social network get stuck in routine behavior at the expense of their own ability to innovate and learn. On the other hand, too little social proximity can be detrimental to the possible establishment of relationships and hamper the ability to learn and innovate interactively because there is a lack of confidence or commitment (Boschma & Frenken, 2010).

2.3.4 Organizational proximity

Entering new relationships for knowledge exchange and innovation is partly dependent on the ability to coordinate the exchange of complementary forms of knowledge, in management or ownership of a variety of actors within and between organizations. Organizational arrangements and agreements, such networks are not only mechanisms that can coordinate transactions but are also vehicles that allow the exchange of knowledge and information. With organizational proximity is therefore meant the same (cognitive) action space in which interactions occur and that connects actors with each other: the extent to which individuals or organizations operate within a similar hierarchical controlling mechanism (Boschma & Frenken, 2010). Organizational proximity refers to a "set of interdependencies within as well as between organizations" (Boschma, 2005, p.65). In other words, organizational proximity is the extent to which relationships are divided into organizational

arrangements and agreements both within and between organizations.

The most important aspect of organizational proximity is the degree of autonomy and the extent to which it can be controlled. As previously argued, knowledge and innovation go hand in hand with uncertainty and opportunistic behavior. To reduce this as much as possible, strong control

mechanisms are necessary in order to secure property rights and to offer an adequate remuneration and / or compensation in exchange for investments in new technologies. It is unlikely that a very large organizational proximity, reflected in a strong hierarchical organizational structure, ensures enough flexibility to encourage new (risky) creative initiatives, knowledge development and innovation. (Boschma & Frenken, 2010). Too much organizational proximity can be harmful for interactive learning, and for the associated innovations and knowledge development, because of the

24 danger of a lock-in and lack of flexibility. On the other hand, too little organizational proximity causes a lack of control to restrain uncertainty and opportunistic behavior.

Organizational proximity can be defined in terms of the difference in routines and excitation mechanisms because they are embedded in a particular institutional context. In this regard, a distinction can be made between organizations with, and without profit purpose (Boschma, 2005). Organizations with a profit purpose will try to keep their own knowledge away from their

competitors, while non-profit organizations, such as universities, have the public task or are more willing to share and exchange knowledge with other organizations. Another distinction is that of frontrunners or leader firms and followers. Frontrunners or leader firms are firms in a cluster that have the power by their size, market position, knowledge and / or entrepreneurial ability, and have the incentive to do investments with positive effects for other companies in the cluster (Langen & Nijdam, 2003). Their presence can ensure that there is more cooperation within the cluster which means that the local buzz increases. Also, these companies can absorb a lot of knowledge form outside the cluster through their size and their position in the market, and thereby contribute to more global pipelines. Similar function for clusters can be fulfilled by 'knowledge brokers’. This involves actors, commercial or otherwise, which act as intermediaries in the exchange of knowledge of business, or actors whom encourage companies to cooperate with academic partners (Atzema et al. 2011b).

2.3.5 Institutional proximity

Where social proximity is defined in terms of social anchored relations at the micro-level,

institutional proximity can be defined as the institutional environment at macro level in the form of "sets of common habits, routines, established practices, rules, or laws that regulate the relations and interactions between individuals and groups" (Edquist & Johnson, 1997, p.46). With institutional proximity is meant the formal and informal institutions such as laws, rules, norms, values, habits and routines at the macro level. Institutions act as a glue for collective action because it reduces

uncertainty and lowers transaction costs. Formal institutions (such as rules and regulations) and informal institutions (such as norms and habits) affect the extent to which individuals or

organizations can coordinate, therewith they can affect the level of knowledge, interactive learning and innovation at the same time. With institutional proximity is to a large extent also meant, the cultural context in which transactions take place.

When such institutions are very similar, it might originate a stable condition for interactive learning and knowledge development and dissemination. On the other hand, institutional proximity can also

25 be a limiting factor between new relationships, in which it hinders interactive learning and

innovations. Too much proximity may result in a lock-in or institutional inertia, resulting from excessive and determining institutions that are not in favor of change, and work in opposition to the development of new innovations. Too little institutional proximity can be a hindrance because a weak institutional context falls short to collective learning behavior and social cohesion. A balanced institutional context consists of institutional stability (to reduce uncertainty and opportunism), openness (to create opportunities for new entrants), and flexibility (to enable experimenting with new institutions).

2.4 The spatial implications of knowledge

The importance of the regional setting for innovation processes depends on the occurring knowledge base in that region and their interplay of knowledge. Lundvall and Borrás (1998) have pointed out that the process of knowledge generation and exploitation requires a dynamic interplay and transformation of tacit and codified forms of knowledge as well as a strong interaction of people within organizations and among them. Thus, the knowledge creative process becomes increasingly inserted into various forms of networks and innovation systems (at regional, national and

international levels).

2.4.1 Industries drawing on different knowledge bases

The next section discusses how different industries rely on different knowledge bases (see figure 3) for activities that are most central to their competitiveness. The place of such an industry in this triangle indicates to what extent the innovative activities rely on the three knowledge bases. So the place of the biotechnology industry indicates that their innovative activities depends mainly on the analytical knowledge base and barely on the synthetic and symbolic knowledge base. Despite the generic trend towards increased diversity and interdependence in the knowledge process, Asheim et al. (2005b) argue that the innovation process of firms and industries is also strongly shaped by their specific knowledge base. A knowledge base refers to the area of knowledge itself as well as its embodiment in techniques and organizations. There are three main categories to be distinguished: ‘analytical’,  ‘synthetic’  and  ‘symbolic’.  

26 2.4.2 Analytical knowledge base

The typology of knowledge bases include and acknowledges the diversity of professional and occupational groups and

competences involved in the production of different types of knowledge (Asheim et al. 2005a). The analytical knowledge base comprises (predominantly scientific)

knowledge that is geared to understanding and explaining features (know-why) of the world. The activities are aimed at a cognitive and rational process of knowledge creation. This means that knowledge inputs and outputs are often codified as they entail knowledge about principles and laws of motion in nature, in the human mind and in society (Lundvall et al. 2002). Such inputs often exist of reviews of existing studies, while knowledge generation is based on the application of scientific principles and methods. Knowledge processes are more formally organized and outcomes tend to be documented in reports, electronic files or patent descriptions (Asheim et al. 2005a). In a regional setting such as BRE, many companies have their own R&D departments and rely on the research results of knowledge institutions in their innovation process. So there is a strong link between university and industry. Typical examples of such industries (see figure 3) are biotechnology and nanotechnology. These industries are also strongly presented in BRE. Another important route of knowledge application is new firms and spin-off companies which are occasionally formed on the basis of radically new inventions or products (Asheim et al. 2005a). Although codification is very frequent, it does not imply that tacit knowledge is irrelevant. As earlier mentioned, properly interpretation and correct use of codified knowledge often requires tacit knowledge.

2.4.3 Synthetic knowledge base

The synthetic knowledge base on the other hand refers to the (predominantly engineering) knowledge involved in the design and construction of solutions (know-how) to human problems which is often instrumental, context specific and practice related (Asheim et al. 2005a). The activities that lead to innovation are mainly aimed at a process of applying existing knowledge and combining knowledge (Asheim & Gertler, 2005). This means that novel knowledge is created in an inductive process of testing, experimentation, computer based simulation or through practical work as its goal

Figure 3: Knowledge bases and industries: empirical example Source: Asheim et al. (2005b)

27 is to solve specific problems coming up in the interaction with clients and suppliers. This way of applied research is often find in plant engineering, specialized advanced industrial machinery and production systems, and shipbuilding (Asheim et al. 2005a). In BRE, where the high tech machinery industry plays a dominant role, a lot of these innovative processes take place in the form of product or process development, such as for example in the automotive industry. In this case the strength of the link between university and industry depends on the form of application. This is usually the case in the field of concrete knowledge application (Asheim et al. 2005a). Such codified knowledge is often embodied in the respective solution or engineering work. However, according to Lundvall et al. (2002) tacit knowledge seems to be more important here than in the analytical knowledge base, due to the fact that knowledge often results from experience gained at the workplace, and through learning by doing, using and interacting. In this case, craft and practical skills are required in the knowledge production and circulation process which are often provided by professional and polytechnic schools, or by on the job training (Asheim et al. 2005a). Their goal of the innovation process is often efficient and reliable solutions, or practical utility of products. Overall, this leads to an accumulating way of innovation, dominated by the modification of existing products and

processes. Since these types of innovation are less disruptive to existing routines and organizations, most of them take place in existing firms, whereas spin-offs are relatively less frequent (Asheim et al. 2005a).

2.4.4 Symbolic knowledge base

Finally, the symbolic knowledge base deals with the creation of cultural meaning through

transmission in an affecting sensuous medium. Activities in this knowledge base are related to the aesthetic attributes of products, to the creation of designs and images, and to the economic use of various cultural artifacts (Asheim et al. 2005a). There is a visible trend of the increasing significance of these activities, looking at the development of cultural industries such as media, advertising, design or fashion (Scott, 1997). These activities are innovation- and design-intensive since a crucial share  of  work  is  dedicated  to  the  ‘creation’  of  new  ideas  and  images  and  less  to  their  physical   production process (Asheim et al. 2005a). However this study is focused on innovation networks within the high tech sector in the Eindhoven region. Since the high tech sector in BRE contains mainly interlinked manufacturing industries with physical production processes, this knowledge base

becomes less relevant to analyze in this particular study. Therefore this knowledge base will not be further discussed nor included in the analysis.

2.5 Conceptual model

28 and  the  ‘learning  region’,  space  and  knowledge  plays  an  explicit  role  in  innovation  processes.  The interpretation of the role of space and knowledge for innovative networks can be accessed on the basis of two central approaches. The first one concerns the notion of proximity on the regional innovation clusters and networks in the region. The second one relates to the knowledge bases and their socio-spatial implications in innovative activities between the key actors involved in the high-tech sector: knowledge institutions, firms and public/governmental institutions. These approaches describe some features which appear to be well calibrated for innovation clusters analysis, especially in high-tech or science-based sectors. For the conceptualization of the spatial organization of

innovation strategies, the next step is to discover the way the notion of proximity, and the

knowledge bases are reflected in innovation strategies. Behind this conceptualization lies of course

the  challenge  of  defining  ‘innovation  strategies’  and  intertwine  it  with  the  two  approaches.  In   addition, the link between the two approaches will be made. The schematic figure of this conceptual model that will unravel the spatial organization of innovations strategies is shown below.

2.6 Innovation strategies

The exchange of knowledge can be considered as an interactive social process based on types of interaction. These interaction take place within networks between firms and institutions. Whether firms participate in networking depends on their innovation strategy (Atzema et al, 2011a). An innovation strategy consists of several stages (invention, development, adaption, marketing) of the innovation process as shown in figure 5. Companies can then focus on internal or external resources. Some firms focus solely on internal resources and don’t participate in networks, they use a

29 alone strategy. A second type of firms focuses on local external resources and uses a 'local buzz strategy'. This means that new knowledge is gained from cooperation with partners located in the same region. The outcome can be cost advantages due to shorter distances, but also gain of

information through face-to-face (f2f) contact between firms. According to Chesbrough et al. (2006) a lot of studies have emphasized the role of formal ties of regional networks in innovation at the organizational level. Such ties are

often  a  visible  part  of  the  firm’s   strategy. However, the role of informal ties is less well understood. These ties reflect an unmeasured confound, which often arise from formal alliances or from casual encounters. Finally, the global pipeline strategy is distinguished for firms that focuses on global external

resources. These firms have their Figure 5: Innovation strategies. Source: Visser & Atzema (2008)

cooperation especially outside their own region. If certain specialized knowledge is needed which is lacking in the region, this strategy has its advantages. The exchange over longer distances usually involves specialized knowledge which is lacking in the region and is important for the innovation process. (Atzema et al, 2011a).

2.7 Reflection of the notion of proximity on innovation strategies

In case of a local buzz strategy, spatial proximity is significant because there is a bigger probability that entrepreneurs or employees of organizations located in the same region meet each other f2f. Nevertheless, this assumption is increasingly questioned (Malmberg & Maskell, 2002). The

relationship between spatial proximity, knowledge and innovative behavior of firms is likely to be more complex. Firstly, it is not necessary to be located close to each other in order to meet on a regular basis. The increased mobility of people and developments in information and

communications technology have made it increasingly easier to meet each other regularly over long distance, meaning that spatial proximity is losing its significance. Firms are more able to have their cooperation with other firms outside the region. Such cooperation establishments are reflected in global pipeline strategies.

It is in the first place important that actors in a cluster can learn from each other and share the same knowledge base (analytical or synthetic). When such cognitive proximity can be reached in the region,

30 actors are likely to choose a local buzz strategy because this is the most inexpensive way of

knowledge exchange. However, it often occurs that certain specialized knowledge is needed, which is lacking in the region. This means that innovation (and cognitive proximity) is reached by a global pipeline strategy, since the relationship with these firms outside the region exists of exchange of knowledge which is compatible.

For the exchange of knowledge, there must be a certain sense of trust between the companies to make cooperation possible (social proximity). It is for firms more easy to build up contacts with organizations in the region then with organizations located further away. This means that a local buzz strategy is preferable when actors give preference to social variables such as reputation, trust and friendly contacts. Because when potential partners see each other on a regular basis, it might originate a certain degree of trust, which makes it more likely that they exchange knowledge (Granovetter, 1985). However, the importance of frequent personal encounters will probably decrease as the firm gets older (Malmberg & Maskell, 2002). New firms that are drawing on a synthetic knowledge base often have many contacts with customers or suppliers because they still have to win their trust. But once this has been established, firms will probably seek other

opportunities to exchange knowledge, because organizing multiple meetings is quite expensive, at least in time. Such other opportunities in winning trust could concern cooperation with firms outside the region (global pipeline strategy). This would mean a shift of partners to stay in social proximity, because personal ties like trust, facilitates knowledge exchange between actors , even if they are not in the same region.

From the perspective of organizational proximity, difficulties are more likely to arise in extra regional partnerships than in partnerships between actors located in the same region. This connection is especially relevant in high technology industries and more radical forms of innovation (innovations new to the sector, as opposed to innovations new to the firm), since in these contexts strong control mechanisms are necessary in order to secure property rights and offer adequate remuneration and / or compensation in exchange for investment in new technologies (Scott & Storper, 1992). So this (cognitive) action space in which individuals or organizations operate in a similar hierarchic control mechanism is more likely to be reached in a local buzz strategy. But firms can also develop

relationships which are too strong. A surplus of organizational proximity can lead to situations where companies become less flexible in their actions (Atzema et al. 2011b). These contacts are likely to limit the innovative capacity of the involved parties. If firms only focus on a few business partners, they might have less regard for other external developments. For that reason firms can decide to start collaborations stemming from global pipeline strategies.